to the Burrampooter in Assam, though its course above the plains remains unexplored, bursts on our knowledge with a stream of such capacity as quite justifies the length attributed to it by the supposition that it is identical with the Sanpoo. And there seems little reason to doubt that the two great rivers, Ganges and Sanpoo, rising from the same lofty region, within 150 miles of one another, after diverging to an interval of some 17° of longitude, combine their waters in the plains of Bengal. In Rennel's time, the Burrampooter, after issuing westward from the Assam valley, swept south and south-eastward, and, forming with the Ganges a fluvial peninsula, entered the sea abreast of that river below Dacca. And so almost all English maps persist in representing it, though this eastern channel is now, unless in the rainy season, shallow and insignificant; the vast body of the Burrampooter cutting across the neck of the peninsula under the name of Jenai, and uniting with the Ganges near Pubna (about 150 miles north-east of Calcutta), from which point the two rivers, under the local name of Pudda, flow on in mighty union to the sea. [11] On the other hand, it is curious that Rennel should have misapprehended the true courses of the other Ngari rivers as he has done. The upper streams of both Indus and Sutlej--the one as flowing past Ladakh from the range of Kylass, and the other past Chaprung from the Rakas lake--are represented with a general truth; but instead of tracing them westward to their true debouchments in the Punjaub, under the well-known names just mentioned, the Ganges is made to draw its waters from the combination of these two Tibetan streams, thus acquiring an imaginary extension of several hundred miles.--(See _Memoir of a Map of Hindostan_, 1778, p. 102.)